TRENDS IN AVERAGE INPUT AND OUTPUT TARIFFS:Â The trends in nominal protection rates (NPR) of import categories reveal that in the recent past the average NPR for imported input categories (basic raw materials, intermediate and capital goods) have been declining rapidly while that of final consumer goods remained practically flat if not increased. The net outcome of this process is higher effective protection to domestic producers over time yielding higher profits simply through tariffs and without any improvement in productivity or competitiveness. Moreover, tariffs on final consumer goods (most of which have domestic production) are too high so that they act as implicit ban on imports resulting in lower revenue yields as well.
Figure 2: Trends in Output and Input Tariffs
 Fig.3 Input and Output Tariff Trends for India and Vietnam
Source: UN TRAINS, WITS
Cross-country comparison reveals that this trend of input and output tariffs is unique for Bangladesh (Fig.2). A comparison of the trend in output and input tariffs of India and Vietnam (Figure 3) shows that both rates trended downwards as tariffs were liberalised thus reducing effective rates of protection or leaving them unchanged over time. This is the approach followed by most developing countries over decades as they liberalised trade thus lowering effective protection to import substitute production with a view to improving efficiency and global competitiveness for exportation - as trade theory would suggest. This is also a strategy for elimination of anti-export bias of tariff policy which has yielded good results for countries like S. Korea, Thailand and Malaysia who have been able to transform import substitute industries for export production, or create new export-oriented industries, resulting in substantial increase in export-GDP ratio over time. Vietnam is in the process of emulating the successful exporters.
Can Bangladesh do otherwise? To attain sustained levels of 7-8% GDP growth, Bangladesh must have a dynamic and globally competitive export-oriented manufacturing sector. To continue on a path of sustainable export growth with a diversified basket of goods, Bangladesh would have to restructure its tariff regime in order to gradually phase out effective protection levels and anti-export bias. A possible tariff and protection regime that could be implemented over the 7th Five Year Plan (FY2016-2020) period is charted in Figure 4. In view of the currently high levels of NPR on final consumer goods, the proposed structure calls for gradual but significant reduction of NPR on these goods while making modest adjustments to input tariffs along the way. Ideally, the average NPR of 50% in FY15 will have to be halved by FY20 while average input NPRs decline from 12% in FY15 to around 9% by FY20. This trend in the structure of output-input tariffs seems to be the only way to go if Bangladesh is to undergo transformative change in its structure of production, where production, jobs and income hinge on the success of exports which, by FY2020, could constitute 30% of GDP, from 17% in FY2015. A beginning will surely have to be made in FY2016, the first year of the 7th Plan.
Figure 4: Possible Input-Output Tariff Regime for the 7th FYP
 SUPPLEMNTARY DUTY ADJUSTMENT SCENARIOS: The way the current tariff structure is laid out, a big portion of the protection comes from Supplementary Duties applied to tariff lines representing final consumer goods which comprise the bulk of production of import substitutes in the domestic market. One of the ways to reform the tariff regime would be to scale back SDs or at least apply them on a trade neutral basis, as the new VAT and Supplementary Duty Act of 2012 requires. It is amply clear to any observer of the Bangladesh industrial scene that removing all SDs or making them all trade neutral is impractical. For instance, the political economy of reducing protection levels by converting all SDs trade neutral in one stroke is too constraining because it will raise strong opposition from the import substituting enterprises. Nor is such a move recommended because of its disruptive impact on the manufacturing sector. Only a phased withdrawal of SDs, implying a phased reduction of effective protection over time, is a feasible option. Such a phased adjustment of para-tariffs (SD and RD) is proposed here along with projections of protection and revenue implications. These projections have been done using an econometric model and applying appropriate methodology for simulating import and revenue response to various schemes of para-tariff adjustments.
Hypothetical Scenario 1: This is a simulation based on the tariff regime that prevails in FY2015 arising from the budget announced in June 2014.
Hypothetical Scenario 2: Full application of the VAT and SD Act 2012, i.e. eliminate SD on all tariff lines except 241 tariff lines where SD is applied on a trade neutral basis. SD then would no longer offer any protection (see Protective SD=0 in column 3 of Table 2) becoming only a revenue measure. The sharp fall in average NPR, from 26.7% for FY2015 to 16.98%, is anticipated to be hugely disruptive to the manufacturing sector. Not recommended.
Phased withdrawal of Protective SD: Hypothetical Scenario 3: Baby step in SD adjustment: Leave all SD on imports as is, but make them trade neutral for those tariff lines that represent alcoholic beverages, cigarettes, motor vehicles, arms and ammunitions. There should be little effective opposition to this step.
Hypothetical Scenario 4: This would be the first serious step in scaling down protective SD. Apart from the adjustments in Scenario 4, remove SD (SD=0) on all RMG tariff lines (RMG needs no protection), and reduce SD to 45% or 30% on all tariff lines subject to 60%. Strongly recommended.
The result of the simulation exercise is summarised below (Table 2-3).
Table 2: SD Adjustment scenarios and protection implications
We have presented analyses in previous studies (Sattar 2014. FY2014-15 Budget and Trade Policy) that the tariff adjustments in FY2015 budget were revenue-neutral despite the scaling down of SD in a wide range of products. Actual developments in import and revenue growth show that phenomenon as both import and revenue up to February 2015 showed growth of 14%. Model simulations (Scenario 1), which are slightly different from actuals, for the entire FY2015 year showed higher import growth (18.311%) with revenue growth close behind. Full application of VAT Law (Scenario 2) would stimulate import growth by 37.831% (due to huge reduction in NPR) but revenue would grow only 27.272% due to the substantial elimination of SD on more than 1200 tariff lines (out of 14600 lines subject to SD). Scenario 3 is a minor step in SD adjustment in the sense that there is no reduction of SD rate on any tariff line. Average NPR is lower by about 3% with import and revenue growth almost similar to scenario 1. Scenario 4, which makes a stronger leap in SD reform, reduces NPR by about 5%, stimulates import growth 18.68% while revenue growth is close behind at 16.64%.
Hypothetical Scenario 4 and protection outcomes are illustrated in Fig. 4a-b and Table 4.
The charts show that rather than diverging trend in input-output tariffs, a first step (if adopted in the first year of the 7th Plan) will have been taken in gradual elimination of protective SD and a phased reduction of protection levels. With a marginal reduction in input tariffs, this measure proposes significant reduction in NPR of consumer goods - a measure that will stimulate imports, infuse competition and contribute to efficiency and productivity improvement, all without much loss of revenue. The reduction in SD also covers a small group of consumer products, mainly RMG (which needs no protection), and some others subject to 60% SD. The expectation is that resistance from import substituting industry (ISI) group will be minimal and the reform can be carried through to the next phase in the following years, possibly completing the full application of the Law in about three years.
CONCLUSION: Bangladesh has had a long history of protecting domestic industries against import competition. Though substantial trade liberaliation was launched in the early 1990s, progress in trade openness stalled by mid-1990s, and further liberalisation was gradual, and sluggish at best. Though ostensibly regarded as revenue instruments, para-tariffs (RD, and SD) emerged as major instruments of protection thus slowing the process of liberalisation. SD, though applied under the VAT and Supplementary Duty Act of 1991 (revised 2012), violated the principle of trade neutrality to become nothing more than a protection instrument. Meanwhile, effective protection levels to domestic consumer goods industries started rising, not falling, through the speedy reduction of input tariffs while output tariffs remained steady, or rising. If this process continues for long, the manufacturing sector, on which the government has put heavy reliance for the rise in GDP growth to 7-8%, is unlikely to become globally competitive and dynamic, growing at 10-15% per annum, with a high performing export sector that is diversified beyond RMG. The strategy in the 7th Five Year Plan consistent with a high performing and export-oriented manufacturing sector will therefore have to include a phased reduction of protection. The elimination or reduction of (unlawful) protection through SD could be the first bold step in this direction.
RECOMMENDATIONS:
* Phased rather than one-shot reduction or removal of SD under the new VAT and Supplementary Duty Act 2012.
* Full application of VAT Law, which is supposed to take effect in June 2015, will have disruptive implications for the manufacturing sector with strong resistance from the import substituting producers. It will entail substantial revenue loss for the Government besides significantly reducing protection levels in one go. Therefore, it is not recommended that the Law take full effect in June 2015.
* A more pliable option is recommended. First, the Government should announce that SD as a protective measure will be definitely phased out because it is a violation of the relevant Law permitting its imposition on imports.
* In the first step, SD can be made trade-neutral for all tariff lines covering alcoholic beverages, cigarettes, automobiles, arms and ammunitions. This could even be done in June 2015 without much opposition. In the next phase, protective SD=60% on RMG can be easily removed because RMG sector needs no tariff protection to exist and prosper on the world market. Tariff lines with SD=60% should be reduced to 30%. In the third year, the VAT and SD Law 2012 can be fully implemented leaving only 241 out of the current 1461 SD tariff lines with trade neutral SD. In the final phase, RD could be eliminated without any revenue loss.
* Under the 7th Plan (FY2016-2020), which relies heavily on a high performing export-oriented manufacturing sector to move to the higher growth trajectory of 7-8%, tariff and para-tariff based industrial protection policy will have to be reviewed and rationalised. Otherwise, it is highly unlikely that Bangladesh's manufacturing sector, beyond RMG, can perform its expected role.
Dr. Zaidi Sattar is Chairman of the Policy Research Institute (PRI). Â
Ziaul Ahsan, PRI's Research Fellow, and Ahmed Sadek
Yousef, PRI's Senior Research Associate , Â
assisted in the preparation of the paper. Â
zaidisattar@gmail.com